As it was explained to me the motor does not have the conventional windings.
The motor is the simplest of motors. Around the outside of the motor are a number of electromagnets. The rotor is an extension of the headstock spindle - it looks like a large toothed shaft. When a magnet is switched on, an intense magnetic field is created and the tooth is pulled to line up with the magnet, then the next magnet is switched on, which forces the spindle to rotate.
Each category of motors has its own unique configuration when it comes to the way that the windings are laid out -- and there are almost more types than any of us would ever imagine. It is definitely true that steppers are just about the simplest design that exists which theoretically would make them the lowest cost. But, so many other factors get involved besides design and construction that it is hard to predict cost on that alone. Originally, steppers were used strictly as positioning devices where torque loads were so low that it was hardly a consideration. The main goal was getting from position A to position B with greatest accuracy and minimal settling time. They are still used for that type of application in such things as the read/write heads in optical and hard disk drives. Even in larger industrial servo systems, steppers were only used where working against a torque load was not a requirement.
More recently, improved stepper motors have been used more like small conventional motors in adjustable speed applications. Improvements in controller design and software algorithms were also factors in the steppers working well in more conventional motor applications. There were some bumps in the road. Those who bought the very early DVR lathes are familiar with the harsh vibration from incorrect software set up. I don't recall if the controller itself had to be replaced.
Basically, all motors function as rotating electromagnets around the stationary field that "pulls" the rotor along. The intensity of the magnetic field is one of the factors that determines the mechanical power output of the motor so that is not a unique characteristic of the motor type.
Since it works by magnetic attraction, the speed or position of the shaft is unimportant to its torque. The switching of the magnets is controlled by a microcomputer, using an digital display and soft control pad. The motor is smart - it knows the exact shaft position and is constantly computing where the shaft should be. At 2000 rpm, for instance, it is computing the shaft position 24000 times per minute.
As with any other motor, the torque output is determined by the load up to the torque limit of the motor (known as stall torque). However, steppers are no different than AC induction motors when it comes to torque output. They may seem to be different if the controller limits the speed range to be within the constant torque operating envelope.
I think that too big a deal is made of "constant torque" -- this is true not only of Nova, but probably all of the other lathe manufacturers who offer sensorless vector inverter drives with three phase AC induction motors. Torque is important to an engineer doing load analysis, but it is only peripherally useful to the machine operator since torque by itself does not do any work. You mentioned full torque at zero speed. Since no work is being done at zero speed, the motor efficiency is zero. Inverter drives are also generally set up to also deliver full torque at zero speed, but an AC induction motor running without an inverter will typically output five to six times full load torque at start-up i.e., zero speed stall torque).
It is the controller and not the motor that is smart -- the motor is just the big galoot that does whatever its master says. Three phase AC induction motors with true vector control can also be set up to work exactly as the DVR motor -- that is control both speed and position. I have several Baldor motors and vector controllers that are able to do that. An optical shaft encoder is used on the motor to provide both precise position and speed feedback information. Unless a stepper also users an encoder, it would not be able to ensure precise positioning because the basic operation is open-loop (meaning that the controller assumes the rotor position, but doesn't account for sync slip).
An even "smarter" motor/controller are the BLDC type (brushless DC), but they are so expensive that they can't yet be seriously considered for woodturning lathes.
The motor itself is only 2 pieces.
It is a lot more than two and comparable to a three phase AC induction motor. However, the stepper is easier to construct and the rotor is less complex because it does not have the "cage".
I've had my DVR 3000 since 2003 and I was able to upgrade the electronics to make it equal to the new DVR. And as stated above either lathe in my opinion would be a good choice.
I think that it is an excellent lathe that is especially well made, but I share the same concerns that John Lucas does about long term availability of replacement parts. The Nova design did not solve the problem of low power at low speeds and in fact put themselves at a power disadvantage in that respect by going to a direct drive when compared to belt drive systems. On the plus side, direct drive does make for a more compact headstock. Other issues are not really clear-cut regarding which of the two approaches is better for doing things.
